:max_bytes(150000):strip_icc()/GettyImages-2207400965-85c1ed8d4a114a25b6be78b4dac7cd03.jpg "BlackRock’s CEO Sees ‘Huge Winners and Huge Failures’ Coming in AI")
:max_bytes(150000):strip_icc():format(jpeg)/GettyImages-2207400965-85c1ed8d4a114a25b6be78b4dac7cd03.jpg)
Key Takeaways
- BlackRock CEO Larry Fink this week defended big AI spending, saying “other countries are going to beat” the U.S. otherwise.
- He also linked the technological change underway to the U.S. economy, questioning whether this year’s lack of job growth should be attributed to uncertain policy or labor substitution.
There are worries about AI spending. Sometimes, that means worries that companies developing artificial intelligence capabilities aren’t spending enough.
Larry Fink, chief of BlackRock (BLK), the world’s largest asset manager, yesterday defended spending on artificial intelligence during a wide-ranging interview at the New York Times DealBook Summit. “If we don’t spend enough—faster—on AI, digitization and tokenization, other countries are going to beat us,” he said.
Fink’s comments landed amid ongoing debates about whether the AI sector is getting over its skis, with some likening the enthusiasm around the tech to the Dotcom bubble. While spending in AI development has triggered investor skepticism and dinged shares of the biggest players in the business, including Oracle (ORCL), Microsoft (MSFT), and Amazon (AMZN), Fink said the technological change underway is already visible in the U.S. job market and corporate margins.
WHY THIS MATTERS TO YOU
As Big Tech companies invest more in AI and lower their headcount, corporate execs, analysts and central bank officials are increasingly referencing what is called a “K-shaped economy,” which describes a situation in which high-income earners and certain industries thrive while lower-income households and other businesses struggle.
The CEO of New York-based BlackRock, which managed over $13 trillion in assets as of the third quarter, said that while hyperscaler CEOs “aren’t certain if they’re overspending or underspending,” their conviction of future demand was high and most don’t have the raw processing power needed to power their AI models.
Fink, however, didn’t outright dismiss the possibility of some companies showing disappointing results. “I’m not here to suggest that there’s not going to be some, you know, headline blow-ups,” he said. “There are going to be some huge winners and huge failures.”
The cost of building global data centers, AI infrastructure, and related power supplies could cost over $5 trillion in the years to come, according to JPMorgan analysts. In order to garner a 10% return on modeled AI investments through 2030 would mean roughly $650 billion of annual revenue in perpetuity, they said.
Fink linked the development of AI technology to the K-shaped economy, which describes a bifurcation in recovery where certain industries and segments of the population experience outsize growth while others struggle.
“What I think is happening, is more and more companies are doing more with the same amount of people or less,” Fink said. “This technological change is happening today, but it’s gonna have a profound impact on our economy.”
